Tuberculosis poses a major global public health problem. Nearly one-third of the world's population (approximately 2 billion people) harbors latent TB infections. An estimated 8 million cases of active TB are diagnosed annually and approximately 2 million people die of the disease. Additionally, the pandemic of HIV disease is resulting in marked increases in morbidity and mortality due to tuberculosis. To complicate matters further, surveys in several areas of the world, including the United States, have shown that up to 10% to 15% of new infections are now due to multiple drug resistant tuberculosis (MDR-TB). Recently, a new class of small molecules, beta-sulfonylacetamides, has been shown to have potent activity against slow-growing pathogenic mycobacteria. FAS20013, in particular, has potent bactericidal activity against M. tuberculosis and MDR-TB. The compound targets a mechanism quite distinct from all currently used drugs. Resistant organisms have yet to be induced in the laboratory or encountered in clinical isolates. FAS20013 kills M. tb relatively quickly, including organisms engulfed in macrophages. The compound also kills organisms equally well that have been adapted to an hypoxic existence (the Wayne model of latency) in contrast to INH and rifampin that are essentially inactive. FAS20013 is orally bioavailable and has the potential of becoming an important therapeutic that can be used to treat M. tb and MDR-TB infections alike as well as to "sterilize" tissue lesions including those of latent infections. The goals of this proposal are to: 1) provide a supply of FAS20013 in the highest state of purity possible, 2) confirm and extend the preliminary studies of the key therapeutic properties of FAS20013, 3) define the pharmacokinetic basis for optimal therapy, 4) formulate FAS20013 to enable an optimal in vivo treatment regimen, 5) complete preliminary toxicity studies of the newly-formulated therapeutic and 6) examine the efficacy of the newly-formulated therapeutic in treating active infections with M. tb and MDR-TB in the mouse model, as well as latent infections. These data will describe the feasibility of treating mycobacterial infections in humans by this method.